{"id":"15e885b6-0328-4023-a32a-d6e340dad738","arxiv_id":"1908.08685","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"First experimental demonstration of frequency-dependent squeezing generated from EPR-entangled optical fields, with a stable phase-control scheme for future gravitational-wave detectors.","lead":"This paper demonstrates a tabletop experiment that produces frequency-dependent squeezed light using two entangled light beams, avoiding the need for bulky filter cavities. The work is a proof-of-principle step toward improving gravitational-wave detector sensitivity across a broad band of frequencies.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the detuning-dependent cancellation in Fig. 2 (right column) already controls for the electronic-artifact concern, though a pump-off null test would strengthen the demonstration.","rationale":"The reader's weakest assumption concerns electronic artifacts in the passive combination of the two homodyne readouts. I agree that this is an unverified assumption and that the paper would be stronger with an explicit null test. However, the paper's own three-configuration comparison acts as a partial internal control: the FD rotation appears or disappears as the optical detunings are changed, including full cancellation in the symmetric case, which would not occur if a fixed electronic phase or amplitude imbalance dominated the signal. Thus the concern is real but not load-bearing enough to change the verdict. The central experimental claim is consistent with the theoretical model in the Supplementary Material, the control scheme is demonstrated, and the stated limitations (2 dB squeezing, high loss, 80-second lock, factor-of-two SNR penalty) are acknowledged by the authors. Missing raw data and error bars reduce confidence but do not invalidate the conclusion. I therefore recommend leaving the ACCEPT verdict unchanged.","tokens_in":751,"tokens_out":664,"duration_ms":125984,"concrete_test":"Perform a pump-off null test: block the OPO pump field (or the idler beam before the mode cleaner) while preserving the LO powers, electronic combiner, and CLF lock. Record the combined-output spectrum with the same idler-LO phase ramp used in Fig. 2 and check whether the spectrum stays at the combined shot-noise level at all frequencies. A flat shot-noise result confirms that the FD squeezing requires the EPR-entangled fields; any residual sub-shot-noise or frequency-dependent structure would indicate electronic crosstalk or common-mode technical noise in the combination stage.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the measured sub-shot-noise, frequency-dependent spectrum at the combined homodyne output is due to EPR correlations processed by the detuned test cavity. The most plausible alternative is a frequency-dependent electronic phase or amplitude imbalance in the two homodyne paths, which could produce an apparent quadrature rotation at the passive combiner without any quantum EPR correlations. That would be load-bearing if present, because it would mimic FD squeezing. However, the paper includes a strong internal control: in the right column of Fig. 2 (delta_sig = -delta_idl = 0.5 gamma_tc), the two optical rotations are arranged to cancel, and the FD feature indeed disappears. A fixed electronic imbalance would not be canceled by changing the optical detuning in this way, so the observed FD behavior tracks the cavity response rather than the electronics. The remaining unverified points, such as equal-gain combination, shot-noise calibration, and the absence of an explicit pump-off null measurement, are limitations on confidence, not demonstrated errors. They do not invalidate the central claim, but they prevent independent verification from the text alone.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an experimental demonstration of frequency-dependent (FD) squeezed vacuum generation using Einstein-Podolsky-Rosen (EPR) entangled fields produced by a non-degenerate optical parametric oscillator (OPO). The signal and idler fields are reflected from a detuned test cavity, spatially separated by a mode cleaner, detected by two balanced homodyne detectors, and electronically recombined. The measured combined noise spectra show frequency-dependent quadrature rotation consistent with the theoretical model for different cavity detunings, including a cancellation case (δsig = -δidl = 0.5γtc) where the FD feature disappears. The authors also demonstrate a coherent locking field (CLF) control scheme that stabilizes the squeezing angle. The total detection loss is estimated from the data to be 47 ± 4% and cross-checked with a power-dependent measurement to be 49 ± 2%.","tokens_in":10232,"tokens_out":4832,"duration_ms":52234,"significance":"If the result holds, the experiment validates a filter-cavity-free method for generating frequency-dependent squeezing, which is of direct interest for future gravitational-wave detectors. The paper's strengths include a complete quantum Langevin model in the Supplementary Material, an independent cross-check of the loss estimate, a detuning-cancellation control that argues against frequency-dependent electronic artifacts, and a scalable locking scheme. The observed squeezing depth of about 2 dB is modest, but it is sufficient to demonstrate the frequency-dependent rotation, which is the central claim.","major_comments":[],"minor_comments":[{"comment":"The measured noise spectra in Fig. 2 are shown without error bars, and the method used to calibrate the shot-noise level and to subtract electronic dark noise is not described in the main text; please add error bars and a brief calibration description so the reader can assess the significance of the 2 dB squeezing feature.","section":"Fig. 2"},{"comment":"The paper states that the signal and idler local oscillator powers are 'tuned to ensure that the readouts are combined equally,' but it does not describe how this equal-gain combination was verified; please add a sentence explaining the verification method (e.g., a known coherent signal or a calibration of the electronic combiner).","section":"Main text, measurement description"},{"comment":"Equation (1) is presented as the noise variance 'scaled to the combined shot noise limit,' but the connection to the factor-of-two normalization used in Eqs. (15)-(16) of the Supplementary Material is not explicit; please clarify that the variance is normalized to the sum of the two individual shot-noise levels.","section":"Methods, Eq. (1)"},{"comment":"The sentence 'The two homodyne readouts are combined passively with an electronic combiner' does not specify whether the combination is a sum or a difference of the two photocurrents; please state this explicitly, as the sign convention affects the interpretation of the squeezing angle.","section":"Main text, paragraph after Fig. 1"}],"recommendation":"accept","confidential_remarks":"This is a well-executed experimental Letter that convincingly demonstrates the EPR-based approach to frequency-dependent squeezing. The detuning-cancellation control in Fig. 2 is a strong internal check against electronic artifacts, and the loss is cross-checked independently. The missing error bars and some calibration details are minor presentation issues that do not undermine the central claim. I recommend acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a clean proof-of-principle of the Ma et al. 2017 EPR scheme for frequency-dependent squeezing, and the internal control in Fig. 2 (right column) is the strongest single piece of evidence that the effect is optical rather than electronic. I think the reader's ACCEPT verdict is right.\n\nWhat's new: first demonstration of FD squeezing generated from EPR entanglement rather than a filter cavity, plus a coherent locking field control scheme that stabilizes the readout angle. The measured spectra follow the theoretical model with only the total detection loss as a free parameter, and that loss is checked independently from squeezing-vs-pump-power. The cancellation case, where detuning the signal and idler oppositely makes the FD feature disappear, is a good control: a fixed electronic phase/frequency artifact would not vanish under that change of optical detuning. That addresses the main alternative explanation directly.\n\nSoft spots: the observed squeezing is only about 2 dB, the total loss is high (47-49%), the experiment runs at MHz sideband frequencies rather than the audio band where GW detectors operate, and the stable lock only lasted ~80 seconds. The authors state all these limitations themselves, so there is little to add. I would have liked error bars on the noise spectra and access to the raw traces; the paper says the dark noise was subtracted but does not show error estimates. A pump-off (or lock-off) null trace would also strengthen the demonstration, though I do not think it is required given the right-column cancellation. The theoretical model in the supplement is standard Langevin/input-output and looks internally consistent; no circular fitting beyond the loss.\n\nOverall, this is an honest, well-documented experimental letter that makes a believable proof-of-principle claim. It is not a deployed detector upgrade, but it is exactly the kind of feasibility demonstration that deserves peer review. I would accept it, asking for error bars or raw data as a minor revision. Citation pattern is appropriate – the Ma et al. proposal and Brown et al. are credited, and the filter cavity context is fairly presented.\n\nRecommendation: send to a serious referee. This is a good reading-group paper too.","headline":"First experimental proof-of-principle of EPR-based frequency-dependent squeezing; the detuning-dependent cancellation control makes the central claim credible, with limitations honestly stated.","tokens_in":10748,"tokens_out":1909,"would_cite":true,"duration_ms":19332,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["42.50.Lc","04.80.Nn"],"model":"deepseek-v4-flash","headline":"This paper demonstrates that frequency-dependent squeezed vacuum can be generated with EPR-entangled fields, using a detuned test cavity to rotate the idler quadrature and a coherent locking field for stable control, offering a…","keywords":["frequency-dependent squeezing","EPR entanglement","gravitational-wave detectors","quantum noise","squeezed vacuum","filter cavity","coherent locking","homodyne detection"],"falsifier":"Lock both signal and idler fields onto the test cavity resonance ($\\delta_{sig}=\\delta_{idl}=0$) and repeat the combined measurement: the model predicts no frequency-dependent quadrature rotation, so any residual FD-shaped spectrum would indicate that electronic or other unbalanced effects, not the detuned cavity, are generating the apparent rotation. A second check is to introduce a calibrated frequency-dependent electronic phase into one homodyne readout while the cavity detuning is zero; if the apparent FD squeezing tracks this injected phase, the readout combination alone can mimic the effect.","tokens_in":9865,"feed_emoji":"⚛️","tokens_out":7505,"duration_ms":65398,"temperature":0.7,"pith_summary":"This paper reports a tabletop demonstration of frequency-dependent (FD) squeezed vacuum made from Einstein-Podolsky-Rosen (EPR) entangled fields, rather than by sending ordinary squeezing through a narrow-band filter cavity. The signal field is kept on resonance with a test cavity while its entangled idler partner is detuned; the detuned cavity gives the idler sidebands a frequency-dependent phase rotation, and recombining the two homodyne readouts conditionalizes that rotation onto the signal, producing squeezing whose quadrature angle rotates with frequency. The measured spectra match the theoretical model and show 2 dB of squeezing below shot noise, with the depth limited mainly by photodiode efficiency. A coherent locking field stabilizes the squeezing angle for about 80 seconds. The significance is practical: it removes the need for a separate suspended low-loss filter cavity in gravitational-wave detectors, at the cost of a factor-of-two signal-to-noise penalty relative to filter-cavity schemes.","feed_headline":"EPR states make frequency-dependent squeezing, no filter cavity","feed_subtitle":"A detuned test cavity rotates the squeezed quadrature; a locking-field control holds the angle steady.","key_machinery":"The load-bearing element is the reflection transfer matrix of the detuned test cavity. In the two-quadrature basis, this matrix has off-diagonal terms proportional to the detuning parameter $\\Delta_{\\gamma_{tc}}$ and to the cavity susceptibility $\\chi_c$, so an input quadrature couples into the orthogonal output quadrature with a frequency-dependent strength. By making the idler field experience this rotation while the signal does not, and then electronically combining the two homodyne readouts, the experiment turns the correlated EPR sidebands into a squeezed state whose squeezing angle rotates with frequency. The control mechanism is the phase-modulated coherent locking field injected into the OPO, which generates error signals that lock both the pump phase and the two local-oscillator phases.","core_discovery":"The central claim is that FD squeezing can be generated and controlled through EPR entanglement plus conditional measurement. In the experiment, an optical parametric oscillator produces signal and idler vacuum fields at frequencies separated by 851 MHz. The signal is resonant with a 0.5 m test cavity while the idler is detuned by roughly one cavity linewidth; the detuned reflection rotates the idler quadrature by an amount that depends on measurement frequency. The two reflected fields are separated by a mode cleaner, measured by independent homodyne detectors, and recombined electronically. The combined spectrum exhibits the expected quadrature rotation: when both fields are detuned equally the whole state rotates to the orthogonal quadrature near the cavity linewidth; when only the idler is detuned, the conditional readout produces a frequency-dependent squeezing angle; and when the detunings are opposite, the rotations cancel and no FD squeezing appears. The demonstration includes a coherent locking field that locks the readout angle, holding the squeezing angle stable for about 80 seconds before slow drift. The authors conclude the technique is a feasible, scalable route to FD squeezing for gravitational-wave detectors, avoiding a separate filter cavity at the price of a factor-of-two signal-to-noise reduction.","pith_inferences":["The passive equal-power electronic recombination is not the optimal Wiener filter described in the proposal, so the full broadband quantum-noise reduction predicted for a detector would require a frequency-dependent weighting of the idler readout; the tabletop result validates the mechanism, not the complete gain.","The conditional-measurement idea is general: the same signal-idler separation and recombination could synthesize user-specified quadrature-frequency responses in other precision measurements, such as optomechanical or atomic sensors, by choosing the digital filter that combines the readouts.","Because the observed depth is dominated by known loss, a straightforward upgrade path is to use high-efficiency photodiodes and cavity impedance-matching; the model's quantitative predictions at higher pump power would be a direct test.","The mode cleaner's role as a frequency-selective beamsplitter could be replaced by a fully digital separation if the two fields are detected together, but this is an engineering extension not pursued in the paper."],"forward_implications":["Gravitational-wave detectors could obtain broadband squeezing-angle rotation by using the signal-recycling cavity or another existing resonator as the filter element, deleting the separate filter cavity and its length stabilization.","The coherent-locking-field control scheme should transfer to a kilometre-scale interferometer, since the locking reference tracks the OPO pump phase rather than relying on a cavity with a 50 Hz linewidth.","Increasing the photodiode quantum efficiency above 80% and reducing mode-matching loss would directly convert the observed 2 dB into much deeper frequency-dependent squeezing, according to the paper's loss model.","Operating with opposite-sign detunings cancels the quadrature rotation, which the paper shows can be used to avoid unwanted rotation in detuned signal-recycling interferometers.","The factor-of-two signal-to-noise penalty relative to filter-cavity schemes is the design trade-off; the paper's demonstration makes this trade explicit for detector planning."],"supporting_citations":[{"why":"Defines frequency-dependent squeezing as the way to beat the Standard Quantum Limit over a broad band and frames the filter-cavity strategy this experiment replaces.","marker":"[2]"},{"why":"Proposes generating FD squeezing with EPR entanglement instead of a filter cavity; the experiment is a direct verification of this proposal.","marker":"[21]"},{"why":"Extends the EPR proposal to a detuned signal-recycling interferometer, motivating the detuning configurations tested here.","marker":"[22]"},{"why":"Introduces the Standard Quantum Limit that sets the measurement target the FD squeezing is meant to surpass.","marker":"[1]"},{"why":"Establishes that quantum noise enters through the dark port and that squeezed vacuum injection reduces shot noise, the conceptual basis of the measurement.","marker":"[7]"},{"why":"Quantifies how optical losses degrade squeezed states in filter cavities, supplying the motivation for avoiding a filter cavity.","marker":"[16]"},{"why":"Assesses realistic filter-cavity performance, against which the EPR approach is compared as an alternative.","marker":"[17]"},{"why":"Supplies the coherent control technique on which the locking-field phase stabilization scheme is based.","marker":"[27]"}],"fun_headline_variants":["EPR states create frequency-dependent squeezing sans filter cavity","EPR entanglement enables frequency-dependent squeezing without extra cavity","Squeeze angle tuned by EPR states and conditional measurement","Frequency-dependent squeezing from EPR without filter cavity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion depends on the assumption that the detuned test cavity is the only thing making the squeezing direction change with frequency, and that the electronics that combine the two measurements do not themselves add a frequency-dependent imbalance that mimics the effect.","fun_headline_variants_meta":{"raw":{"variants":["EPR states create frequency-dependent squeezing sans filter cavity","EPR entanglement enables frequency-dependent squeezing without extra cavity","Squeeze angle tuned by EPR states and conditional measurement","Frequency-dependent squeezing from EPR without filter cavity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000549,"raw_usage":{"total_tokens":2594,"prompt_tokens":888,"completion_tokens":1706,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":504,"completion_tokens_details":{"reasoning_tokens":1642}},"tokens_in":504,"tokens_out":1706,"duration_ms":13130,"temperature":1.0,"reasoning_tokens":1642,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:31:32.786212+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Lock both signal and idler fields onto the test cavity resonance ($\\delta_{sig}=\\delta_{idl}=0$) and repeat the combined measurement: the model predicts no frequency-dependent quadrature rotation, so any residual FD-shaped spectrum would indicate that electronic or other unbalanced effects, not the detuned cavity, are generating the apparent rotation. A second check is to introduce a calibrated frequency-dependent electronic phase into one homodyne readout while the cavity detuning is zero; if the apparent FD squeezing tracks this injected phase, the readout combination alone can mimic the effect.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines frequency-dependent squeezing as the way to beat the Standard Quantum Limit over a broad band and frames the filter-cavity strategy this experiment replaces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Proposes generating FD squeezing with EPR entanglement instead of a filter cavity; the experiment is a direct verification of this proposal."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Extends the EPR proposal to a detuned signal-recycling interferometer, motivating the detuning configurations tested here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the Standard Quantum Limit that sets the measurement target the FD squeezing is meant to surpass."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that quantum noise enters through the dark port and that squeezed vacuum injection reduces shot noise, the conceptual basis of the measurement."},{"cited_title":", author Miller, J","cited_arxiv_id":null,"evidence_quote":"Quantifies how optical losses degrade squeezed states in filter cavities, supplying the motivation for avoiding a filter cavity."},{"cited_title":", author Barsotti, L","cited_arxiv_id":null,"evidence_quote":"Assesses realistic filter-cavity performance, against which the EPR approach is compared as an alternative."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the coherent control technique on which the locking-field phase stabilization scheme is based."}],"review_version":1}